1 / 43

UK ATLAS Upgrade project Report to STFC OsC

UK ATLAS Upgrade project Report to STFC OsC. Craig Buttar University of Glasgow Swindon 24/3/11. Outline . Information in presentation is complementary to the OsC report Heritage: ATLAS Construction and Tracker Upgrade project LHC upgrade programme (presented at sLHC2011 meeting)

said
Download Presentation

UK ATLAS Upgrade project Report to STFC OsC

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. UK ATLAS Upgrade projectReport to STFC OsC Craig Buttar University of Glasgow Swindon 24/3/11

  2. Outline • Information in presentation is complementary to the OsC report • Heritage: ATLAS Construction and Tracker Upgrade project • LHC upgrade programme (presented at sLHC2011 meeting) • ATLAS Upgrade programme (presented at sLHC2011 meeting) • Highlights of work from 2010-11. 1st ATLAS Upgrade OsC

  3. UK ATLAS Upgrade project • 13 Universities • STFC/RAL PPD • STFC TD RAL/ATC • Three year project • April 2010-March 2013 • R&D for ATLAS Upgrade programme • Maintain UK leadership • Builds on heritage of current ATLAS development and construction andTracker Upgrade project 1st ATLAS Upgrade OsC

  4. Heritage: ATLAS construction • Last ATLAS “construction phase” oversight committee signed off on all deliverables • Semiconductor tracker (assembled in UK) • Level 1 calorimeter trigger (led by UK) • High level trigger & central DAQ (major aspects led by UK) • Completed within £82M budget • No call on £6M contingency 1st ATLAS Upgrade OsC

  5. ATLAS running and Upgrade roles • ATLAS is fully operational and is producing physics results • Continue major roles in ATLAS operation and physics including • Deputy Spokesperson: Dave Charlton • SCT PL: Steve McMahon • L1CALO PL: Steve Hillier • Deputy Chair Publications Committee: Tony Doyle • Trigger co-ordination group: John Baines, Simon George • Physics group convenors • Upgrade roles • Upgrade coordinator: Phil Allport • Thermal management: Georg Viehhauser • TDAQ Upgrade coordinator: Norman Gee • Software coordinator: Jeff Tseng • Phase-I TDAQ coordinator: Nikos Konstatinidis 1st ATLAS Upgrade OsC

  6. Heritage: Tracker Upgrade project • In the last 6 months of the last grant we achieved our final deliverables, these have lead seamlessly into the ongoing work plan, but the final highlights could be described as: • Completion of the first multi module object in the world and its readout through HSIO. • Completion of the first thermo-mechanical stave (full size) with 24 working heat modules, fully instrumented and connected to a representative cooling plant (CO2) • Completed readout and powering systems for staves and stavelets, that are compatible with 1. and 2. and have since been used extensively to characterise both objects • In short, the UK completed the last phase of the project by producing a 4 module full electrical prototype, and a 24 module mechanical one. • The following 6 months (bridging and the last few months) has been spent understanding these objects and refining the design to enable the next step to the first complete electrical stave 1st ATLAS Upgrade OsC

  7. Heritage: Tracker Upgrade project Thermo Mechanical stave about to be tested Module under test Serially powered stavelet under test 1st ATLAS Upgrade OsC

  8. ATLAS Organisation 1st ATLAS Upgrade OsC

  9. LHC and ATLAS Upgrade programme

  10. How the lumi might evolve in LHC Prudent assumptions, of September 2010: Better than expected LHC behaviour is not yet integrated. It is assumed to saturate at design luminosity of 1.e34. Today we may assume 1.7-2 e34 ! The new shutdown plan (approved 31 January 2011) not yet integrated (shutdown in 2013) Better performance may push the integrated lumi to 300 fb-1 before 2020. L.Rossi sLHC2011 1st ATLAS Upgrade OsC

  11. LHC Upgrade: The goal • The main objective of HL-LHC is to implement a hardware configuration and a set of beam parameters that will allow the LHC to reach the following targets: • A peak luminosity of 5×1034 cm-2s-1 with levelling, allowing: • An integrated luminosity of 250 fb-1 per year, enabling the goal of 3000 fb-1 twelve years after the upgrade. This luminosity is more than ten times the luminosity reach of the first 10 years of the LHC lifetime. L.Rossi sLHC2011 • allow design for lower peak L, less pile up • less peak heat deposition ( a factor 2 may be the diff 1st ATLAS Upgrade OsC

  12. Summary of ATLAS/LHC Upgrades Time-line Start of LHC 2009 Run 1: 7 TeV centre of mass energy, luminosity ramping up to few 1033 cm-2 s-1, few fb-1 delivered LHC shut-down to prepare machine for design energy and nominal luminosity ATLAS Phase-0 upgrade 2013/14 Run 2: Ramp up luminosity to nominal (1034 cm-2 s-1), ~50 to 100 fb-1 Injector and LHC Phase-I upgrades to go to ultimate luminosity 2017 or 18 Run 3: Ramp up luminosity to 2.2 x nominal, reaching ~100 fb-1 / year accumulate few hundred fb-1 Phase-II: High-luminosity LHC. New focussing magnets and CRAB cavities for very high luminosity with levelling ~2021/22 Run 4: Collect data until > 3000 fb-1 Nigel Hessey ATLAS Upgrade sLHC2011 ILC, High energy LHC, ... ? 2030 1st ATLAS Upgrade OsC

  13. Why Upgrade ATLAS? There are many important results that will benefit from a much bigger data set Technology improves, and we can build better performing detectors now Detectors age, especially with accumulated radiation damage, and need replacement: better to plan ahead and replace with the best allowed by technology The LHC will improve, in particular delivering higher instantaneous luminosity than ATLAS was designed for: needs higher granularity detectors to maintain performance It takes a long time to install new detector elements, which has to be done with the LHC off ATLAS will take maximum advantage of all LHC shut-downs to make the best possible detector It takes many years to research ideas, design upgrades, and build them, especially new inner trackers, hence the work started several years ago Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  14. 2013/14: Insertable B-layer There are several reasons to bring forward the IBL installation into the 2013 shut-down: -- It improves the physics performance of ATLAS, even at low luminosity -- Less risk/tooling needed to install when the pixel detector is removed for nSQP installation -- Lower radiation environment in 2013 compared to 2017/18 ATLAS IBL project will accelerate to install in 2013/14: -- very challenging, but looks possible x2 improved background rejection New front-end readout chip FE-I4 delivered and performing very well Nigel Hessey ATLAS Upgrade sLHC2011 WP5: Pixels 1st ATLAS Upgrade OsC

  15. 2017/18:Phase I: Conditions and trigger Phase-I Task-Force in place to optimise ATLAS Phase-I upgrades Options under study given here; what cannot be done in Phase-I will move into Phase-II About 9 months shut-down for installation of new elements Conditions after: Peak luminosity increasing to 2 x 1034 cm-2 s-1 Total integrated luminosity before Phase-II 3-400 fb-1 Trigger: Beef up processors and data links for extra data rate Bring in “topological” triggers – the ability to look at 2 or more trigger objects at L1 e.g. isolated muon = muon far from any jets Missing Et significance trigger WP6: L1 Calorimeter WP7: L1 Track trigger WP8: HLT Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  16. Level-2 fast track finder, FTK (before the SD) WP8: HLT Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  17. ATLAS Changes for Phase-II, HL-LHC, around 2021 Conditions: Peak luminosity 5 x nominal with luminosity levelling: 200 interactions per bunch crossing to be disentangled 3000 fb-1 good data on tape: Very big increase in integrated luminosity --> high radiation dose to detectors 18 month shut-down Most of ATLAS can remain: Magnets, most of muon and calorimeter systems. Changes summary: Trigger and DAQ: significant changes needed Several new muon chambers needed - to be evaluated with experience New calorimeter readout for higher granularity trigger information Changes in LAr End-cap calorimeter New inner detector Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  18. Trigger at sLHC Need to maintain low thresholds on leptons (~20 GeV), missing ET, and forward jet trigger for the physics programme. Events are ~5x bigger, storage and bandwidth limit us to same final event rate as now (~200 events/s - but we are investigating implications of raising this) So must reject 5x as many events of 5x the size – challenging Single particle rates at low pT are too high; raise single object thresholds but maintain low thresholds in combination with other features. Main improvements: Muon trigger – increase the sharpness of the threshold at higher pT (40 GeV/c) Calorimetry: read out all data, full granularity, and build trigger off-detector - allows better particle ID at L1 Longer L1 latency (6 or 12 ms cf 3 now), allowing more processing for combined objects Possible inner-tracker track-trigger at L1 Precise trigger scheme will evolve as the physics priorities and detector capabilities become better known In addition, data storage and transfer bandwidth need beefing up WP6: L1 Calorimeter WP7: L1 Track trigger WP8: HLT Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  19. Inner Tracker: Completely New Hit rates in current inner tracker: Current pixel B-layer becomes noticeably inefficient at 2x1034 cm-2 s-1, significantly so at 3x1034 cm-2 s-1 SCT: some regions cannot readout events above 2.5x1034 cm-2 s-1, due to optical data-link bandwidth TRT occupancy becomes very high, although it still helps even at 3 x 1034 cm-2 s-1 Radiation damage: SCT designed for 700 fb-1: above that, progressively worse inefficiency and other problems Pixel B-layer considerably less New technology: New electronics (130 nm and smaller CMOS) allows lower power and smaller chip sizes; for pixel read-out chip, less inactive area New bump-bonding and chip thinning allows cheaper, thinner pixels New cooling and carbon support structures can allow lower radiation lengths Multiplexing, e.g. local powering schemes, and CO2 cooling allow reduced material budget Conclude: ATLAS needs an all new Inner Tracker at Phase-II Higher granularity detectors to keep occupancy down Base-line is an all-silicon tracker: pixels and micro-strips Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  20. Strawman Layout of New ATLAS Inner Tracker 4 layers of pixels to larger radius than now 3 double-layers of short strips (SCT region) 2 double-layers of long strips (TRT region) Approx. 400 Million pixels (cf 80 Million now) Approx. 45 Million strips (cf 6.3 Million now) WP9: Computing Implemented in Geant, including realistic service material, to study performance and look at optimisations Inner Tracker Sub-committee set up to further improve on this: number of layers, length of barrel, conical end-caps, maintenance Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  21. Micro-strips: sensors n-in-p Bias Ring Guard Rings SiO2 n+ p-spray/stop p- bulk p+ Al Choose n-in-p (cfp-in-n now) Faster signal collection, cheaper production than n-in-n, does not need full depletion Successful production ATLAS07 sensors at Hamamatsu - irradiation tests and prototyping made WP2: On-detector (Tracker upgrade project) Nigel Hessey ATLAS Upgrade sLHC2011 Neutron irradiation results show S/N worst case is 10:1 in strawman layout after 3000 fb-1 with a safety factor 2 1st ATLAS Upgrade OsC

  22. Microstrips: Modules and staves Hybrid with front-end chips glued directly to sensor Sensor glued to cooled mechanical support - “Stave” Staves arranged in cylinders Stave can reduce material and helps assembly schedule by avoiding bottle-neck at module mounting on cylinders WP2: On-detector WP3: off-detector WP4: Mechanics (Tracker upgrade project) Mini-stave built: very good and uniform front end performance (noise, gain, pedestal, threshold); low dead channel count 1st ATLAS Upgrade OsC

  23. Track Trigger at L1 Several ideas for implementing a track trigger at L1. Wanted: high-PT (~20 GeV) leptons. ATLAS EM calo has good identification, allowing a two-stage trigger approach: Calorimeter or muon system identifies a candidate high-PT lepton and gives region-of-interest Inner tracker modules in that region are read-out, and hardware track finders confirm presence of track with matching momentum RoI is a few % of modules so small increase in bandwidth needs --> very little increase in material Needs additional data stream in FE chip and a lot more study, but encouraging so far WP7: L1 Track trigger Alternatively, measure track angle to radial direction at outer edge of inner tracker - look for near radial tracks Either with paired silicon layers or GasPix detector with 10 mm drift gap Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  24. UK-ATLAS Upgrade project results from 10/11 • Tracker Upgrade • WP2: On-detector systems • WP3: Off-detector systems • WP4: Mechanics • WP5: Pixels • Trigger Upgrade • WP6: Level-1 Calorimeter trigger • WP7: Level-1 Track triger • WP8: High Level Trigger • Software, Simulation and Computing (WP9) 1st ATLAS Upgrade OsC

  25. WP2: On Detector Systems Delivers ASICs and Hybrids to the Module programme; Modules, Tapes and On-Stave Interfaces to the Stavelet and Stave programmes ABCN-25 BCC • 2.2 ASICs • Digital Design • Evaluation • Wafer Test • 2.4 Modules • Build Process & Tooling • Evaluation • Skill Transfer • 2.6 On-Stave Interface • Design • Evaluation • 2.5 Tapes • Layout & Manufacture • Flying Probe Machine • Evaluation • 2.3 Hybrids • Layout & Manufacture • Die Attach Tooling • Evaluation • Skill Transfer • 2.1 Strip Sensors • Layout • Evaluation Picture shows Serially Powered Stavelet 1st ATLAS Upgrade OsC

  26. WP2: Recent Deliverables For the DC-DC Stavelet First DC-DC Test of Stave Module First DC-DC Stavelet DETAIL • Module (WP2) • Cu Plated Shield (WP3) • DC-DC convertors (CERN) • Custom Module Frame PCB (WP2) • Results in agreement with SP module • provided adequate shielding used • Stavelet Core (WP4) • DC-DC Power Tape (WP2) • Misc. Support PCBs (WP2, not shown) • Construction of 4 modules in progress (WP2) 1st ATLAS Upgrade OsC

  27. WP3 Off-Detector SystemsSerial Power, DAQ, optics and test facilities for the stave prototyping programme Serial Powering (M3.1, M3.3): Constant current supply tested, tuned, and more to come Power protection hardware prototype to be implemented in 130nm ASIC Passive Optics (M3.6): Facility for cold irradiation of candidate fibres being deployed at CERN Building 180 at CERN being established as a test facility for stave(lets) (M3.2) • Data Acquisition (M3.4, M3.5, M3.7): • UK provides software, firmware and support for HSIO roll-out to UK and overseas collaborators • UK leads development for tie-in with future on-detector chipset 1st ATLAS Upgrade OsC

  28. WP4: Thermo-mechanical Stave Stave locking mechanisms • Full-scale thermo-mechanical mock-up of stave including; • Carbon-fibre/honeycomb sandwich with embedded stainless steel/Pocofoam cooling structure • 24 thermo-mechanical modules including silicon substrates and copper/kapton hybrids based on emerging design from WP2 • Aim: to measure thermo-mechanical performance on realistic, full-scale object & compare with FEA • Thermal Resistance • FEA: 0.043 C/W • TMS: 0.045 C/W Dial gauges measure deflection Thermal resistance Top: 0.0425 ± 0.0024 °C/W Bottom: 0.0474 ± 0.0030 °C/W All: 0.0449 ± 0.0037 °C/W 28 1st ATLAS Upgrade OsC

  29. WP4: Stave QA • Thermal Testing • IR camera • Good uniformity – no sign of any failures of the pipe to facesheet thermal coupling Dummy stavelets • 3-point bending • Check facesheet-core adhesion • Good consistency • Flatness • CMM touch probe scan • Typically within +/- 0.1mm • New tooling should result in better flatness • Long-term thermal cycling • Chamber constructed • First trials in progress 1st ATLAS Upgrade OsC

  30. WP4: Module Mounting & Stave Containers • Module Mounting (2010) • 24 thermo-mechanical • 4 electrical modules • Plans for 2011 • Extend to full length stave • Develop pick-up tooling • Glue studies • Stave Containers • Stavelet container defined • New Module-mounting frame defined • Manufacture imminent • Scheduled to link in with international stavelet programme 1st ATLAS Upgrade OsC

  31. WP5: Sensors: 6” Micron Pixel Wafer n-in-p, single metal FZ process The 2nd mask will be used to have devices for IBL qualification and PPS studies of radiation hardness and high voltage operations • 5 FE-I4 tiles/wafer • 4 with 8 MPI “style” guard rings • 1 with 6 RD-50 “style” guard rings • 14 FE-I4 single chips/wafer • 2 with RD-50 “style” guard rings (390 mm wide) • 8 with 8 MPI “style” guard rings (390 mm wide) • 4 with 4 MPI “style” guard rings (200 mm wide) • And other test devices • Available, 2nd mask: • 8x300µm thick wafers. • 2x150µm n-in-p wafers • 2x150µm p-in-n wafers • In preparation (ready end of April): • 2x300µm n-in-n wafers • 3x150µm n-in-n wafers • 3D sensors delivered for IBL prototyping and pre-production “RD-50” “MPI” 1st ATLAS Upgrade OsC

  32. WP5: Planar 4-chip module • Use Micron sensors: including 4-chip (quad) sensor • One FE-I4 wafer • Use VTT for bonding • VTT first • Use 6 off existing microns sensors • Note lots of experience so don’t feel they need heavy qualification • Quad and singles • New Micron wafer • Produce Quads for module programme and singles for HV tests, testbeam etc • Mechanicals • Make sensor/detector daisy chain for mechanicals with over hang for wire bonding • From new Micron wafer • Timescales • Quad design ready 12/4/11 • Quad sensors ready 30/09/11 • Mechanical module 02/12/11 • Electrical module 10/08/12 • 3D sensor and module development 1st ATLAS Upgrade OsC

  33. WP6 Calorimeter Trigger Simulation of L1 calorimeter trigger rates at different luminosities and with different thresholds Presented at L1CALO week in Cambridge, this week One possible strawman design for Phase-II L1Calo upgrade consisting of two parts, a fast synchronous level-0 accept followed by an asynchronous level-1 system incorporating higher granularity features. 1st ATLAS Upgrade OsC

  34. L1Track simulation studies • Developed framework for readout modeling studies • To investigate/optimize data formats for readout, bandwidths etc, both for track trigger and normal readout, working together with experts from the Tracker upgrade project • To understand requirements for track trigger and feed into the design “Distance” of Pixel wafer from z=0 % of RoIs containing a barrel Pixel module Due to spread of LHC luminous region in z, modules near z=0 more often inside a Region of Interest. 1st ATLAS Upgrade OsC

  35. L1Track: L1 trigger studies at 2e34 L1_MU20 in barrel • Validation of ATLAS high luminosity simulation • L1 muon turn-on curves show that above 20GeV the current L1Muon system has small discrimination power in pT • Raising pT thresholds cannot provide adequate improvement in L1 rate rejection • Further information required by trigger – track trigger? • Studies being extended to • higher luminosities • tau and electron triggers L1_MU40 in barrel 1st ATLAS Upgrade OsC

  36. WP8 : Performance Benchmarking • Have established test-bed to benchmark performance on MC datasets up to 2x1034 cm-2s-1 • Measured performance: • Efficiency of track reconstruction as a function of luminosity • Execution time as a function of occupancy • Identified areas for optimisation & improvement • New Level-2 Inner Detector Tracking Package created : • Will incorporate optimised L2 components • Allows evaluation of different reconstruction strategies. Level-2 Inner Detector Tracking (MuonRoI) Execution time v. no. spacepoints Level-2 Inner Detector Tracking (MuonRoI) Top events 1034 pileup top events No pileup 1034 pileup 2x1034 pileup 1st ATLAS Upgrade OsC

  37. WP8 & WP9: Use of GPU in the HLT Measure possible speed-up of L2 code on GPU c.f. CPU • Ported HLT code to GPU: • Zfinder • Track Fitter • Measure execution times c.f. CPU • Next steps: • Increase parallelisation of fitter • Add: • Data preparation • Pattern Recognition • => Complete tracking chain on GPU GPU – Fermi ~factor 10 speed up for Fermi GPU Factor 35 speed-up for L2 Zfinder running on GPU GPU time almost flat 1st ATLAS Upgrade OsC

  38. WP9: Athena, Radiation • Work ongoing on putting AthenaMP on Grid • Makes more optimal use of all cores on a node • Big improvement in throughput & memory per process • Test queue at RAL established for ‘whole node’ scheduling • Commissioning studies to finish in July • Radiation Environment • Simulations compared with 2010 data • Reliable 1MeV neutron equivalent damage fluences established • Detailed detector material and services added to FLUKA simulations • Good description for SCT barrel and RadMons (that measure ionizing dose) • Poorer description for SCT endcap inner modules under investigation 1st ATLAS Upgrade OsC

  39. WP9: Simulations etc Framework development • Made robust and adapted to incorporate new geometries • New, radical, geometries implemented • Advances in efficient handling of pile-up • Reduced memory requirement for digitisation • Modifications for Inner Beam layer project • Histograms, including those for Atlfast tuning • Patches to be integrated with the main software release Conical pixel Layout • Studies of current Utopia layout • Help design, test simulation & reconstruction • Fake rates reduced but still significant 1st ATLAS Upgrade OsC

  40. Summary • Changes to LHC schedule have been addressed by ATLASClear programme of R&D established towards TDRs • Representation at highest levels of ATLAS management allows UK to influence and adapt to changes • UK Upgrade programme has adapted to changes in ATLAS programme • UK Upgrade programme is up and running • Results presented at ATLAS Upgrade week at CERN • New results will be presented at ATLAS Upgrade week in Oxford • UK is maintaining and enhancing its leadership within the upgrade programme 1st ATLAS Upgrade OsC

  41. BACK UP SLIDES 1st ATLAS Upgrade OsC

  42. Physics goals of sLHC Main ATLAS Physics goals: Higgs discovery: Mass and understanding electro-weak symmetry breaking Unification of forces, gravity, SUper SYmmetry, extra dimensions New forces (W', Z') Flavour: why 3 families, neutrino mass, dark matter Whatever is discovered at the LHC will need a lot of data to understand exactly what has been discovered: characterising the discoveries. In addition, the sLHC can extend the discovery potential, to higher masses or lower cross-sections. While the LHC aims at ~300 fb-1 per experiment, the sLHC aims for 3000 fb-1 of data, opening up new possibilities for channels limited by statistics at the LHC There are many measurements where extending the LHC data set is important, including: 1. Higgs couplings 2. Triple gauge-boson couplings 3. Vector boson fusion at ~1 TeV 4. SUSY – discovery or spectroscopy 5. New forces: W', Z' to higher limits Nigel Hessey ATLAS Upgrade sLHC2011 1st ATLAS Upgrade OsC

  43. WP2 WP3 WP4 Connections between WP2345 Serial Power systems Sensor Development WP5 WP5 ASICS Power/Signal Tape low mass development mixed Al/Cu Hybrid Manufacture methods Module Manufacture DAQ FEA Mounting Methods Strip Core Fabrication low mass techniques high rigidity WP5 Stave Design Stave Assembly WP5 Optical Fibre study Stave Test Cooling low mass pipes welding Materials properties radiation tolerance Integration WP5 WP5 nb not all activity/links shown

More Related